Hotword-Based Speaker Recognition in Multi-User Environments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In a multi-user speech-enabled environment, existing systems face challenges in accurately identifying the speaker to access and manage user-specific resources, particularly when ambient noise or new configurations interfere with voice recognition.
Innovation Solution
The system employs a hotword for dual purposes as a trigger and for speaker identification, using Mel-Frequency Cepstrum Coefficients (MFCCs) and nearest neighbor analysis to match speaker features, and prompts users for identification if a match fails, associating determined usernames with user accounts for resource access.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional voice recognition systems are used in multi-user environments, then the system can process speech queries, but it cannot accurately identify which user is speaking, leading to incorrect resource access
Solution Approach 1:
The hotword serves dual purposes: it acts as both the trigger to activate the system and as the basis for speaker identification. By using the same word for both functions, the system avoids adding separate identification mechanisms while maintaining accuracy in multi-user environments.
2Reliability
If the system requests user identification in noisy conditions or during configuration, then identification accuracy improves, but user interaction time increases
Solution Approach 1:
The system performs speaker identification on the hotword itself, which is already being processed for activation. This preliminary identification action occurs simultaneously with hotword recognition, eliminating the need for separate identification steps and reducing user interaction time while maintaining reliability.
3Ease of operation
If the system uses only hotword recognition without speaker identification, then the system is simpler to operate, but it cannot securely manage user-specific resources
Solution Approach 1:
The hotword serves dual purposes: it acts as both the trigger to activate the system and as the basis for speaker identification. By using the same word for both functions, the system avoids adding separate identification mechanisms while maintaining accuracy in multi-user environments.
Data Source
AI summary
Systems, methods performed by data processing apparatus and computer storage media encoded with computer programs for receiving an utterance from a user in a multi-user environment, each user having an associated set of available resources, determining that the received utterance includes at least one predetermined word, comparing speaker identification features of the uttered predetermined word with speaker identification features of each of a plurality of previous utterances of the predetermined word, the plurality of previous predetermined word utterances corresponding to different known users in the multi-user environment, attempting to identify the user associated with the uttered predetermined word as matching one of the known users in the multi-user environment, and based on a result of the attempt to identify, selectively providing the user with access to one or more resources associated with a corresponding known user.